DREADDed microglia in pain: Implications for spinal inflammatory signaling in male rats

Abstract

The absence of selective pharmacological tools is a major barrier to the in vivo study of microglia. To address this issue, we developed a G_q- and G_i-coupled Designer Receptor Exclusively Activated by a Designer Drug (DREADD) to enable selective stimulation or inhibition of microglia, respectively. DREADDs under a CD68 (microglia/macrophage) promoter were intrathecally transfected via an AAV9 vector. Naïve male rats intrathecally transfected with G_q (stimulatory) DREADDs exhibited significant allodynia following intrathecal administration of the DREADD-selective ligand clozapine-N-oxide (CNO), which was abolished by intrathecal interleukin-1 receptor antagonist. Chronic constriction injury-induced allodynia was attenuated by intrathecal CNO in male rats intrathecally transfected with G_i (inhibitory) DREADDs. To explore mechanisms, BV2 cells were stably transfected with G_q or G_i DREADDs in vitro. CNO treatment induced pro-inflammatory mediator production per se from cells expressing G_q-DREADDs, and inhibited lipopolysaccharide- and CCL2-induced inflammatory signaling from cells expressing G_i-DREADDs. These studies are the first to manipulate microglia function using DREADDs, which allow the role of glia in pain to be conclusively demonstrated, unconfounded by neuronal off-target effects that exist for all other drugs that also inhibit glia. Hence, these studies are the first to conclusively demonstrate that in vivo stimulation of resident spinal microglia in intact spinal cord is a) sufficient for allodynia, and b) necessary for allodynia induced by peripheral nerve injury. DREADDs are a unique tool to selectively explore the physiological and pathological role of microglia in vivo.

Keywords: AAV; Allodynia; CNO; Chemogenetics; Gene therapy; Intrathecal; Minocycline; Neuropathic pain.

Figure 1. Spinal microglia inhibition via CD68-hM4Di attenuates CCI-allodynia and CD11b upregulation in vivo

(a) Naïve rats were transfected with intrathecal hM4Di or control DREADDs. Two weeks later, CCI was performed on all rats. After a further two weeks, all rats received a single intrathecal dose of CNO (60 μg). Von Frey thresholds were determined prior to CCI (baseline; BL), prior to CNO (0 h; 14 days post CCI), and across a 24 h timecourse after CNO injection. (b) CD11b density was quantified in the ipsilateral dorsal horn of lumbar spinal cords from naïve rats, nerve-injured rats expressing control DREADD and treated with CNO for 3 days (60 μg/day), and nerve-injured rats expressing hM4Di DREADD and treated with CNO for 3 days. Representative images of ipsilateral dorsal horn of lumbar spinal cord are shown for (c) naïve rats (d) nerve-injured rats expressing control DREADD and treated with CNO for 3 days, and (e) nerve-injured rats expressing hM4Di DREADD and treated with CNO for 3 days. Data are presented as mean ± SEM; n = 6/group; *P < 0.05, ***P < 0.001.

Figure 2. BV-2 cell inhibition via hM4Di attenuates LPS- and CCL2-induced inflammation

BV-2 cells were transfected with hM4Di. (a–d) Cells were treated with a concentration range of LPS, together with 0 or 50 μM of CNO. After 4 h, (a) Nos2 or (c) Il1b mRNA was quantified from cell lysates. After 24 h, (b) NO or (d) IL-1β levels were quantified in supernatants. (e–h) Cells were treated with a concentration range of CCL2, together with 0 or 50 μM of CNO. After 4 h, (a) Nos2 or (c) Il1b mRNA was quantified from cell lysates. After 24 h, (b) NO or (d) IL-1β levels were quantified in supernatants. Data are presented as mean ± SEM; n = 5/group; ***P < 0.001.

Figure 3. Spinal microglia stimulation via CD68-hM3Dq induces IL-1-dependent allodynia in vivo

Naïve rats were transfected with intrathecal hM3Dq or control DREADDs. Four weeks later, all rats received a single intrathecal dose of CNO (60 μg). Intrathecal IL-1ra (100 μg) was administered 2 h after CNO. Von Frey thresholds were determined prior to (baseline; BL), and across a 24 h timecourse after CNO injection. Data are presented as mean ± SEM; n = 6/group; *P < 0.05, ***P < 0.001.

Figure 4. hM3Dq stimulation of BV-2 cells induces pro-inflammatory mediators in vitro

BV-2 cells were transfected with hM3Dq, and treated with a concentration range of CNO for 24 h. Supernatants were collected and analyzed for (a) NO, (b) TNF, (c) IL-1β, (d) IL-6. Data are presented as mean ± SEM; n = 5/group; *P < 0.05, ***P < 0.001.